TW201636775A - Power and load management based on contextual information - Google Patents

Abstract

A power context system is described herein that makes decisions related to device power usage based on factors such as location, load, available alternatives, cost of power, and cost of bandwidth. The system incorporates contextual knowledge about the situation in which a device is being used. Using the context of location, devices can make smarter decisions about deciding which processes to migrate to the cloud, load balancing between applications, and switching to power saving modes depending on how far the user is from a power source. As the cloud becomes more frequently used, load balancing by utilizing distributed data warehouses to move processes to different locations in the world depending on factors such as accessibility, locales, and cost of electricity are considerations for power management. Power management of mobile devices is becoming important as integration with the cloud yields expectations of devices being able to reliably access and persist data.

Description

Power and load management based on background information

The present invention relates to power and load management based on background information.

From mobile phones to data centers, energy efficiency has gradually become an important differentiator. Consumers are willing to pay extra for longer-lasting mobile device experience, but consumers are also eager to get improved performance from these same devices. On the other hand, the data center continues to amplify the computational power, but faces thermal limitations on whether it can be efficiently cooled. In addition, the public is increasingly concerned about the use of energy and the impact of energy on the environment. Therefore, for many types of computing systems, efficient use of energy is a higher priority design goal.

As mobile devices become widely distributed, power management and integration with the cloud are becoming increasingly important to provide uninterrupted access and data persistence on mobile devices. Today's mobile devices typically have relatively few power states. The device is not operating at full power, or in standby or idle state and operating at low power. If the battery level is low, the device can perform some steps to reduce power usage, such as dimming the screen. However, such an action can prevent the user from using the device.

The data center faces similar issues. In some cases, an organization operates multiple data centers, and the organization has the flexibility to route operations to more than one data center according to certain criteria. For example, a data center operator can only select the amount of requirements that the data center can handle to the data center because the data center has computational and cooling capacity limitations. The operator can then transfer any overflow request to another data center at another location.

Decisions made in such situations are often not optimal, as complete situational knowledge is not used to make decisions. For example, the data center may not be pre-considered, such as: according to the time of day, the client request is about to drop a lot, so there is no need (or only temporarily needed) offloading work. Similarly, the mobile device may not be pre-considered if the user plans to plug the device into 10 minutes, so even if the existing battery is low, it is still sufficient to allow the user to use the full device performance. The results of such decisions are not as good as those that can be expected to be achieved by smarter devices, or by using a comprehensive set of background knowledge available to the devices operating in the environment.

This document describes a power background system that makes decisions regarding device power usage based on factors such as location, load, available alternatives, power cost, and bandwidth cost. The system incorporates background knowledge about the context in which the device is used in the context. Using the location background, the device can determine which programs to transfer to the cloud and between applications based on how far the user is from the power source. Load balancing and decision to switch to power save mode. As the cloud becomes more and more used, by using a decentralized data warehouse to move programs to different locations in the world based on factors such as accessibility, locale, and power costs, Load balancing is a consideration for power management. Because mobile devices are prevalent and integrated with the cloud to create device requirements that reliably access and store data, power management of mobile devices is becoming increasingly important. A device that makes intelligent decisions based on the context of the environment and the available resources in the system to manage power and offloads operations using more processor resources to the cloud to save power is more useful to the user. As a result, power background systems can provide lower data center costs, extended mobile battery power, and more satisfying user experience.

This [invention] is provided to introduce some conceptual choices in a simplified form, and the concepts will be further described in the following [Embodiment]. This invention is not intended to identify key features or essential features of the claimed subject matter, and is not intended to limit the scope of the claimed invention.

100‧‧‧Power background system

110‧‧‧Power status components

120‧‧‧Load sensing components

130‧‧‧Power cost components

140‧‧‧Location awareness components

150‧‧‧Decision Engine Parts

160‧‧‧Load Transfer Parts

170‧‧‧Power reserve parts

180‧‧‧Application Communication Unit

210-250‧‧‧Steps

310-350‧‧‧Steps

Figure 1 is a block diagram showing the components of a power background system in an embodiment.

2 is a flow chart illustrating the processing of a power background system that manages power in an arithmetic device based on location in an embodiment.

Figure 3 is a flow chart illustrating the processing of transferring the computational load from one computing device to the power background system of another computing device in one embodiment.

This document describes a power background system that makes decisions regarding device power usage based on factors such as location, load, available alternatives, power cost, and bandwidth cost. This system is more intelligent than previous systems because it incorporates background knowledge about the situation in which the device is used in the context. Using the location background, the device can make decisions about determining which programs to transfer to the cloud, load balancing between applications, and switching to power save mode based on how far away the user is from the power source (where "far" can be a geographic distance , or until the user is expected to be close to the available power source). For example, if the user is using a mobile device with low battery power to execute an intensive application, but depending on the user's location, the user will arrive at home very quickly (the device can be charged at home), then the system can proceed Less power-saving action than when the user is moving away from the power source. Similarly, if the data center server is processing the load during a power cost spike in one region and the other data center is available in another region during the power cost off peak, the server can transfer the load to use. Another data center with less power costs. The availability of such alternatives and background information are factors that make decisions based on the system.

The power background system makes decisions related to power costs. For mobile devices, the power cost stops the device. For the data Heart, power consumption represents money. One job of the system is to use a variety of background information and externalities to determine what to do to optimize power costs. If a cheap power source (for example, in a data center) is about to be available, or the power cost is about to decrease, the decision result can be continued. Otherwise, the decision result can be transferred to another "node" where the power cost is lower. The phone can transfer the workload to the data center, and the data center can be transferred to another data center (or phone).

The device can also make intelligent decisions about which data link to make based on location and historical knowledge/data. For example, if the user is close to the home/office, the device may pause the download on the cellular network and wait until a higher bandwidth Wi-Fi network connection is available. Some devices also limit the download size on some networks, and wait to allow larger files to be downloaded. If the device knows that a better link will be available soon, the device can put the download into the queue to begin downloading when the appropriate network is available.

As the cloud becomes more and more used, load balancing is achieved by leveraging decentralized data warehouses to move programs to different locations in the world based on factors such as accessibility, regionality, and power cost. For power management considerations. Power management of mobile devices is becoming increasingly important as mobile devices are prevalent and integrated with the cloud to create device requirements that reliably access and store data. A device that makes intelligent decisions based on the context of the environment and the available resources in the system to manage power and offloads operations using more processor resources to the cloud to save power is more useful to the user. Transfer can also occur in other directions, The cloud offloads work to other devices' networks (such as the SETI program). As a result, power background systems can provide lower data center costs, extended mobile battery power, and more satisfying user experience.

For mobile devices, effective application and resource power management helps extend battery power to the device. The application can require higher computing power when the mobile device is plugged in or fully charged. However, when battery power is low, continued use of high operating power will quickly drain the battery, causing possible data loss or device usage disruption. Therefore, depending on the amount of power available to the mobile system, it can be allowed to execute with applications of different power amounts. Event-based subsystems help balance the computational load between applications based on available power. Events allow the application to communicate the amount of computational power and memory resources required, as well as the amount of resources available to the application. The system monitors event and load balancing between applications to stop applications that use more computing resources at low power and redistribute resources between applications when more resources become available.

The power background system can use the geographic location of the device and expect that until the power source is available, determine which applications are allowed to execute and which applications are closed. When the power is low, the application of the intensive operation is usually first turned off. However, if the device can reach the home within a few minutes using the location background of the device and the target decision device (there are available inexpensive power sources in the home), the device may not need to reserve power. Therefore, an application that uses more computing resources can continue to execute, although the system power state is low. Use location background to allow loading Dynamically make intelligent decisions and anticipate whether the device needs to retain power. In addition, background information about the availability of alternative resources, such as the cloud that can be accessed by the device, may allow the device to offload the computationally intensive work to the cloud while the device is put to sleep or allow the device to execute other applications. Depending on the location of the device, the user may be able to receive the result of the reduced work on another device located at a subsequent location (eg, a user's working desktop) or on a mobile device connected to the power source.

The power background system may also use heuristics based on the sensing environment to determine when to access a wide variety of device subsystems based on current power/location/mode of use. System calls to the core use more computing resources than other calls. A handshake between the core and the sensor can be used to manage the power between applications. For example, while the device is in a low power state, the system can allow a small subset of low power sensors to operate, and if the sensor detects something significant, the system can wake up the device and enter a higher power state. To deal with more complex events for a short time. In some cases, the sensor can provide a lower fax, lower power mode in which some of the sensing comprehension can be traded off with lower power usage. This provides good enough information to detect if it is worthwhile to wake up more sensors to use more power costs to provide a higher read fax.

Program state transitions to the cloud can be used to offload work using more computing resources, provide storage in the cloud, or transfer programs to another device to continue processing (such as transferring from a phone to a computer). cloud The end can act as a virtualized environment with extensions of mobile devices, and the cloud can be seamlessly integrated with the mobile system for load balancing. Programs such as background services executing on the system can be switched to a virtualization program executed from the cloud.

As cloud computing becomes more prevalent, power management in data centers has become an important issue. Load balancing procedures in a decentralized data center can be based on electricity costs. The program can be moved to a different location depending on where the cost of the power is cheaper, or the program can be scheduled so that execution is not completed during the peak time. For example, the daily batch processing routine can be delayed until the peak time to allow for cheaper power costs, more processing while the external temperature is low (and thus the cooling cost of the data center is reduced), etc. . Because cloud computing is a virtualized environment, bandwidth and security are important issues to deal with. Depending on regional, weather and geopolitical factors, data can be restricted to certain data warehouse locations. For example, some agencies may be restricted from storing certain information in a particular country, and the transfer of data centers located in their countries is prohibited. The frequency of access to the data can also be considered as a consideration for where to push the data. Frequently accessed data is kept nearby, while less frequently accessed data can be moved to locations that are less accessible (and cheaper or underutilized).

Figure 1 is a block diagram showing the components of a power background system in an embodiment. System 100 includes a power state component 110, a load sensing component 120, a power cost component 130, a location awareness component 140, a decision engine component 150, a load transfer component 160, and a power retention component 170 and application communication component 180. Each of these components is described in further detail herein.

Power state component 110 maintains information that describes the current power state of the computing device. For example, component 110 includes knowledge of whether the mobile device is being actively used, in a sleep state, dimmed backlight/screen, and the like. For the data center, component 110 includes performance counters, operating system power states, device state knowledge, and the like. Power state component 110 provides information to system 100 to make a decision as to whether to enter a different power state, and how much power is currently being used by the device. Based on the requirement to receive more or less power, system 100 can use this information to transfer the device to a higher or lower power state. For example, if the user desires to execute an intensive application and the device is in a sleep state, system 100 can determine if the device is to be woken up, or system 100 can delay the user request to another time (or delay to another device or Cloud).

Load sensing component 120 senses one or more load requirements caused by activity on the computing device. The load can come from client requirements, interactive users, application background work, operating system services, batch processing, or any other work that uses resources on the computing device. The load sensing component 120 provides information for other times when the load begins, when there is a future load desired, when the load is completed, and the possible transition from one load level to another. System 100 uses this information to select between available power states of the device and determines if an alternate device will be used to handle all loads by transferring the load to the alternate device Or some load. For example, a data center server that is experiencing high load (and corresponding high power cost) can transfer some of the load to another data center to complete.

The power cost component 130 determines the current power cost and future power cost of the computing device. Component 130 can receive information from a utility or manager that identifies peak times and off-peak times or other power cost levels throughout the day. Component 130 can also receive information identifying the geographic location of the computing device, daily temperature information, and other information that affects power costs. For the mobile device, component 130 can cooperate with position awareness component 140 to account for the power cost in the current location relative to the power cost in the desired subsequent location. Costs may include financial costs and differences in the effort to get more power, the use of batteries in mobile devices, and the use of power lines (eg, the former may be limited, while the latter may not be limited) and the like.

Position awareness component 140 determines the current location of the computing device, and location awareness component 140 determines one or more possible subsequent locations of the device. The component 140 can access a global positioning system (GPS) receiver hardware, cellular network information, Wi-Fi information, or other means for triangulating or determining the location of the device installed in the device. Location data. The data center can also detect the location through clear configuration information or through externally available information such as GPS, Wi-Fi or other signals. Some of today's data centers are even mobile and contain data centers that are housed in containers and moved by vehicles. Additionally, component 140 can access an electronic calendar associated with the device user to determine that the device is not The location at the time of the appointment. Component 140 can also track historical information, and component 140 tends to determine the user's habits and where the user can take the device at any particular time. For example, if the current location of the device is near a location where the user brought the device to and from the same time every day, component 140 may determine that the device is about to be at a normal target. The target can be the user's home or office, and the target can be a location where the device typically has access to additional power, communication, or processing resources. This information is considered by the decision engine component 150 as a factor to determine how to manage the device.

Decision engine component 150 receives information from other components, and decision engine component 150 makes a decision regarding the computing device power state and whether to transfer the load to other devices. Decision engine component 150 can use a variety of heuristic algorithms, settings, strategies, and other information or methods to make decisions about how to manage the device. In some cases, managers develop power strategies for devices that are related to an organization or enterprise. For example, the policy may instruct decision engine component 150 to maximize battery power when the user is away from the power source (eg, by offloading work or entering a lower power state), and the policy may instruct decision engine component 150 when the user is near the power source Maximize processing performance. For the data center, the policy can instruct component 150 to minimize power cost, heat boost, and/or bandwidth cost by, when possible, offloading work to the off-peak data center. Decision engine component 150 motivates load transfer component 160 and power reserve component 170 to make decisions and perform other actions.

The load transfer unit 160 transfers the arithmetic load from the arithmetic unit to the auxiliary arithmetic unit. In some cases, the computing device is a mobile device The auxiliary computing device is a data center or a cloud server, and the data center or the cloud server can perform processing on behalf of the mobile device. In other cases, the computing device can be the data center server itself, and the secondary computing device can be another data center server, a decentralized mobile device network, a home server, or any other computing resource. Transferring the load means finding a good processing location to handle the load at low cost, low power, or for other purposes. The transfer may include the transfer of the overall program, or the transfer may include portions of the program that can be cut and distributed.

The power retention component 170 transfers the computing device from the current power state to another state, or the power retention component 170 prevents the transition from the current power state. In some cases, component 170 may override the preset behavior of the device to transfer power states to prevent diversion. For example, this decision can be based on location, based on subsequent expected access to additional power. In other cases, component 170 may, depending on instructions received from decision engine component 150, include instructions to transition the device from a higher power state to a lower power state based on instructions of location, power cost, available alternatives, and the like (and vice versa) ).

The application communication component 180 optionally provides an event-based communication facility between applications operating on the computing device to coordinate power requirements and power availability on the computing device. Much of the work performed by the computing device is optional, or much of the work performed by the computing device can be postponed to another time without obstructing the purpose of the work. For example, the computing device may perform batch processing, system maintenance, or other work that is routine but does not have a specific deadline. Therefore, if the system maintenance worker To tell other applications that the system maintenance work wants to execute, and no other application responds, the system maintenance work can be performed without hindering other applications. On the other hand, if another application indicates that another application is performing intensive processing, or another application wants to use the last amount of power available, maintenance can delay its work to allow power or resources. Can be used in other applications.

The computing device implementing the power background system on the computing device may include a central processing unit, a memory, an input device (such as a keyboard and an indicator device), an output device (such as a display device), and a storage device (such as a disk drive or other non- Volatile storage media). The memory and storage device is a computer readable storage medium, and the computer readable storage medium can be encoded by computer executable instructions (eg, software) that implement or enable the system. In addition, the data structure and message structure can be stored or transmitted via a data transmission medium, such as a signal on a communication link. A wide variety of communication links can be used, such as the Internet, local area networks, wide area networks, point-to-point dial-up links, mobile phone networks, and the like.

Embodiments of the system can be implemented in a wide variety of operating environments, including personal computers, server computers, handheld or laptop devices, multi-processor systems, microprocessor-based systems, and programmable consumption Electronic products, digital cameras, network personal computers, mini computers, large computers, distributed computing environments including any of the above systems or devices, set-top boxes, system-on-chip (SOC), and the like. The computer system can be a mobile phone, a personal digital assistant, a smart phone, a personal computer, a programmable consumer electronic product, a digital camera, and the like.

The system can be illustrated in the context of a general computer executable instruction, such as a program module, which can be executed by one or more computers or other devices. In general, program modules contain routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In general, the functionality of the program modules can be combined or dispersed as desired in a wide variety of embodiments.

2 is a flow chart illustrating the processing of a power background system that manages power in an arithmetic device based on location in an embodiment.

Beginning at block 210, the system detects the intensive use of power on the computing device. For example, a user may execute an application or request information, an application may require execution of a background or other processing, a service of the operating system may require execution of system maintenance, and the like. The system can receive information from the application or other sources describing the resource requirements for the work, such as throughput, memory, bandwidth, or other resources that will be used. Based on this information, the system can determine whether work is allowed to be performed, or the system can determine whether to perform work remotely on another device.

Continuing at block 220, the system determines the current power state of the computing device. For example, the power state can include the voltage level of the processor, the memory bank being turned on and off, the brightness of the screen or the state of the backlight, whether one or more sensors and communication hardware are powered, and the like. The system collects power status information from the device, and the system can query the application to determine the power state that is currently desired to complete the existing work. Depending on the power state, the system can determine the new power state to which the device is to be transferred, or the system can determine Does the job of adding new requirements overload the device or use too much power.

Continuing at block 230, the system determines the proximity of the device location to additional power and/or computing resources. The location may include a GPS location, a conceptual location (eg, 5 miles away from home), an address, or help the system determine power availability to determine if the current power level of the device is sufficient to complete the work occurring there and determine if other devices can Other information that handles work more efficiently (eg, based on lower power costs at the location of other devices).

Continuing at block 240, the system identifies one or more alternate devices to perform the requested work. For example, the system can determine whether there is an available link to the data center or cloud, whether other devices are using Bluetooth or other communication protocols for proximity, whether the device is docked or tethered with another computing device, and bandwidth cost (eg, According to the action data plan or the data center bandwidth cost) and so on. The system may also require information describing the availability and power costs of the alternate computing resources, so that the decision engine can determine where to place the processing work to manage efficiency and cost.

Continuing at block 250, the system performs a power-based action to manage the computing device based on the determined device location and other received information. The actions may include transferring the current power state of the device to another power state, preventing power state transitions of the device, transferring required work or other loads performed on the computing device to one or more alternative computing devices, and the like. System selection will provide the best possible experience for one or more users of the computing device while efficiently using power and tubes The action of cost. The actions performed may be based on one or more policies defined by the system administrator or specified by the device user to meet the specific goals of the institution or user. After block 250, the steps end.

Figure 3 is a flow chart illustrating the processing of transferring the computational load from one computing device to the power background system of another computing device in one embodiment.

Beginning at block 310, the system receives a request to perform an operation that will generate an operational load on one or more resources of the computing device. For example, the user can launch the application, the data center can sample the server program to handle user requirements, and the like. The system can use heuristic algorithms or received information to determine the level of the required load and identify which resources the load will use. The system uses this information and other information to determine whether to perform the load locally, remotely, or a combination of the two.

Continuing at block 320, the system determines the location of the computing device, and the location determines the availability of additional power and/or computing resources. The location may include a GPS location, a conceptual location (eg, 5 minutes from home), an address, or a help system determining power availability to determine if the current power level of the device is sufficient to complete the work occurring there and determine if other devices are more Other information that efficiently handles work (eg, based on lower power costs at the location of other devices). In some cases, the location information includes the latency and bandwidth used to determine the location of the remote location to complete the work, the expected time to complete the work (when the system is performing work remotely (relative to local)), and the like.

Continuing at block 330, the system identifies one or more alternative devices to perform the requested work. For example, the system can determine whether there is an available link to the data center or cloud, whether other devices use Bluetooth or other communication protocols for proximity, whether the device is docked or wired to another computing device, bandwidth cost, and the like. The system may also require information describing the availability and power costs of the alternate computing resources, so that the decision engine can determine where to place the processing work to manage efficiency and cost.

Continuing at block 340, the system determines a power cost for performing the required load locally on the computing device, and a power cost for the one or more identified alternatives. Power costs can include monetary costs, the heat load generated at each location, the availability of other devices to other jobs, and the like. The system can also determine the time of day in each location and the power cost based on the time of one of the days determined. For example, many electric utility companies charge different rates for peak and off-peak times of the day. Because the data center can be located in different time zones, the data center in the peak cost location can be able to transfer at least some of the work to the data center in the off-cost location.

Continuing at block 350, the system determines if the required load is to be performed on the local computing device, or the system determines whether to transfer at least some of the required load to the identified alternate computing device. The system can make decisions based on location information and environmental factors, and location information and environmental factors will change as the computing device arrives at a subsequent location. The system may also be determined based on power cost, bandwidth cost, or other information describing the load on the local computing device and one or more of the identified alternate computing devices. example For example, if the battery of the mobile device is low, or when it is predicted that all available battery power is needed when receiving a new job, the device can transfer the work to the cloud for execution, and the device can reserve the battery power of the mobile device to Other jobs. As another example, a data center in a high-power cost area can delay work until a lower power cost time of the day, or the data center can transfer work to a lower cost data center. The system selects the most efficient location to perform each job based on the additional information available (not considered by previous systems). After block 350, the steps end.

In some embodiments, the power background system operator has a software license, or other agreement with the operator (or other individuals that influence the decision made by the system). For example, operators of cellular networks can define policies that specify power settings using Wi-Fi, 3G, 4G, or other sensors and communication hardware. For some user data schemes, the operator may choose to use the setting that minimizes the communication hardware cost for the operator, although for other data schemes, the operator can optimize the battery power or throughput of the device. In some cases, the operator may provide additional plans to maximize the quality of service (QoS) experienced by the device user.

In some embodiments, the power background system can manage communication costs by selecting a particular communication hardware. While processing and power costs have been used in the examples herein, users are faced with other types of costs. For example, users who use the 3G solution provided by the cellular network operator (which plans to include a limited amount of data before levying additional costs) may want to use Wi-Fi links whenever possible (when available) Time). In some cases, depending on the location information, the device may wait until the user returns home, and be able to perform some work that would otherwise have been performed using the 3G data plan resources via Wi-Fi (and therefore without using 3G data plan resources). The mobile device may include a setup phase during which the user provides such preferences, or may be provided by an operator or other individual, such as an information technology (IT) person at the location where the user is employed.

In some embodiments, the power background system takes into account the cost associated with transferring the load to the remote location. For example, a transfer can include the use of bandwidth for transmitting work and receiving results. Some work can also involve user input or other communications in the middle of a variety of tasks. Based on this information, the latent cost or bandwidth cost of the transfer load may not be expected, even if the transfer may be beneficial to other factors (such as maximizing local device battery power). Therefore, the system can take a variety of concerns and completion goals as a consideration to determine the appropriate location to perform each load.

In some embodiments, the power background system includes historical data storage, and the historical data storage includes historical information observed by the system. The system can use learning or other techniques to observe trends, and the system can enhance the decision making process based on past user habits, trends aggregated from a large number of user subjects, or other information. The system can store information over time and the system can use this information as feedback to make the system better make decisions. In some embodiments, the system provides a report to the manager who can modify the system based on the current decision to meet the suitability of power, cost, and other goals.

It will be appreciated that the specific embodiments of the power background system have been described herein for purposes of illustration, and various modifications may be made without departing from the spirit and scope of the invention. Therefore, the invention is not limited, except as limited by the scope of the appended claims.

100‧‧‧Power background system

110‧‧‧Power status components

120‧‧‧Load sensing components

130‧‧‧Power cost components

140‧‧‧Location awareness components

150‧‧‧Decision Engine Parts

160‧‧‧Load Transfer Parts

170‧‧‧Power reserve parts

180‧‧‧Application Communication Unit

Claims (20)

A computer implemented method for managing power in an computing device according to a location, comprising the steps of: detecting a work of intensive use of power on the computing device; determining a current power state of the computing device; determining the a location of the device and a proximity to additional power and/or computing resources; identifying one or more alternative devices for performing the requested operation; and determining the location of the device and other received information based on the determined location Performing a power-based action to manage the computing device; wherein the foregoing steps are performed by at least one processor. The method of claim 1, wherein the step of detecting the operation of the densely used power comprises the step of determining an application executed by a user of the computing device. The method of claim 1, wherein the step of detecting the intensive use power comprises the step of receiving information from an application or other resource indicating one or more resource requirements for the work. The method of claim 1, wherein the step of detecting the operation of the densely used power comprises the steps of: determining whether the work is allowed to be performed locally or whether the remote is at another This work is performed on the device. The method of claim 1, wherein the step of determining the current power state comprises the step of determining at least one of: a voltage level and frequency of a processor, one or more memory banks that are turned on and off. , a screen brightness, a backlight state, and whether one or more sensors, external devices, and communication hardware are powered. The method of claim 1, wherein the step of determining the current power state comprises the steps of: collecting power status information from the device and querying an application to determine a power state that is currently desired to complete the existing work. The method of claim 1, wherein the step of determining the current power state comprises the step of determining a new power state to which the device is to be transferred. The method of claim 1, wherein the step of determining the location of the device comprises the steps of: determining a power availability and determining whether the current power level of the device is sufficient to complete an additional power source to be reached at the user. A job that occurred before a location. The method of claim 1, wherein the step of determining the location of the device comprises the step of determining if the other device can process the job more efficiently. The method of claim 1, wherein the step of identifying the replacement device comprises the step of: determining a link to Whether it is available in one of the detected data centers or the cloud. The method of claim 1, wherein the step of identifying the alternate device comprises the step of requiring information describing the availability and power cost of the alternate computing resources. The method of claim 1, wherein the step of performing a power-based action comprises at least one of: transferring the current power state of the device to another power state, preventing the power from the device State transitions and transfer of the required work or other load performed on the computing device to one or more alternative computing devices. The method of claim 1, wherein the step of performing a power-based action comprises the step of selecting an action that will provide the user (or users) of the computing device with a good experience At the same time, power and management costs are used efficiently. A computer system for power and load management based on background information on an computing device, the system comprising: a processor and a memory, the processor and the memory configured to execute software instructions implemented in the following components; a power state component that maintains information describing a current power state of the computing device; a load sensing component that senses the Calculating one or more load requirements generated by activities on the device; a power cost component that determines a current power cost and a future power cost of the mobile computing device; a position awareness component that determines the operation a current location of the device, and the location awareness component determines one or more possible subsequent locations of the device; a decision engine component that receives information from other components and makes a decision relating to the operation a power state of the device and whether to transfer the load to the remote device; a load transfer component that transfers an operational load from the computing device as indicated by the decision engine based on a determined monetary power cost And a power retaining component that transfers the computing device from the current power state to another power state or prevents transitioning from the current power state based on the location information. The system of claim 14, wherein the power cost component cooperates with the location awareness component to clarify the power cost in the current location relative to the power cost in the one or more desired subsequent locations. The system of claim 14, wherein the location awareness component accesses an electronic calendar of the user associated with the device to determine that the device is in place during a future appointment Reasonably located. The system of claim 14, wherein the location awareness component tracks historical information, and the location awareness component tends to determine the user's habits and determines where the user may bring the device to at any particular time. At the office. The system of claim 14, wherein the decision engine component receives a power policy from a manager that determines one or more targets that the decision engine is attempting to satisfy. The system of claim 14, the system further comprising an application communication component, the application communication component providing an event-based communication facility between the applications operating on the computing device to coordinate the Power requirements and power availability on the computing device. A computer readable storage device containing instructions for controlling a computer system to transfer an operational load from one computing device to another, wherein the instructions cause a processor to perform an action when executed The actions include: receiving a request to perform a job that will generate an operational load on one or more resources of the computing device; determining a location associated with the computing device, the computing device determining Availability of additional power and/or computing resources; Identifying one or more alternative devices for performing the required work; determining a power cost on the computing device and for performing the required load on the one or more identified devices identified And automatically determining that the required load is to be performed on the local computing device, or transferring at least some of the required load to the identified one of the alternative computing devices.